BIS July Papers Trace How Financial Risk Travels
Six new BIS working papers this month examine how stress moves between sovereigns, banks, non-banks, currencies and firms.
Six BIS working papers published in July converge on a single question: where does financial risk go once it leaves its point of origin, and who ends up holding it. The month's research spans domestic credit rules in Colombia, sovereign risk pricing across emerging markets, exchange rate pass-through, dollar-linked stablecoins, and the financing structure of the AI investment boom. The connective thread is transmission: how a shock in one corner of the financial system shows up in another.
The clearest statement of that theme comes from "The evolving nexus: sovereigns, banks and NBFIs," by Stefan Avdjiev, Bryan Hardy and Maximilian Jager, which argues that the sovereign-bank doom loop identified after the eurozone crisis has grown a third leg. As sovereign debt has climbed across major economies, non-bank financial institutions have built up a large enough footprint in government bond markets that stress can now pass between sovereigns, banks and NBFIs in either direction. That widening web of exposure sets up two more targeted papers in the same cluster. "Geopolitical risk and emerging market sovereign risk premia," by Fredy Gamboa and José Vicente Romero, finds that geopolitical threats move sovereign credit default swap and EMBI spreads even before they escalate into acts, using panel data across 13 emerging markets back to 2005. That points to anticipation effects in how sovereign credit gets priced. "Assessing the effects of recent provisioning rules on consumer credit allocation in Colombia," by Diego Cuesta-Mora, Fredy Gamboa and Camilo Sanchez-Quinto, works the other end of the chain. After the Superintendence of Finance of Colombia raised provisioning requirements for long-term consumer loans in January 2023, the paper traces how credit institutions adjusted lending strategies under the resulting profitability pressure.
A fourth paper turns from credit risk to price formation. "What drives exchange rate pass-throughs? Evidence from a non-parametric method," by Emanuel Kohlscheen and Aaron Mehrotra, applies random forests to four decades of data across close to a hundred countries to rank the factors that determine how much a currency move feeds into consumer prices. The method is chosen because pass-through relationships are highly non-linear and resist standard regression.
Aaron Mehrotra appears again alongside Boris Hofmann and Jan Paulick in "Dollarisation and monetary control: what lessons for the rise of stablecoins?", which treats dollar-pegged stablecoins as a new variant of an old phenomenon: deposit dollarisation. Drawing on data covering more than 130 economies, the paper compares stablecoin inflows against the historical pattern of foreign-currency deposit dollarisation, framing the question as one of monetary control in emerging and developing economies.
The month's outlier by subject, "The AI investment race," by Phurichai Rungcharoenkitkul, still fits the transmission theme in its own way. It models AI infrastructure spending as a contest in which firms racing for a limited number of dominant positions over-commit resources, financed heavily by debt and circular equity arrangements between firms. Calibrated to balance sheet and deal data, the model estimates over-investment at roughly 1.5 times the efficient level, rising to about three times when demand is assumed less elastic, and traces how a network of financial exposures could carry stress from one firm to others.
Across the six papers, the categories used to describe financial risk (sovereign, bank, non-bank, currency peg, corporate balance sheet) turn out to be porous to each other. Each paper isolates a different channel, but the sum is a research month asking the same question from six angles: once stress starts moving, what stops it.